Barbara Scalvini at the Faculty of Science and Technology at the University of the Faroe Islands has, in collaboration with the Centre for Language Technology, developed a user friendly website where you can use the Speech Recognition model created by Carlos Mena. This speech recognition model is mainly built on the BLARK by the Ravnur Project.
Please note that this is a trial page, so the speech recognition takes its time. You can expect that it takes as long to process a sound clip as the sound clip is long. This speech recognizer can only process one sound clip at a time, so if someone else is using the speech recognizer exactly when you press start, you will end up in a queue. But your speech recognition will start as soon as the other speech recognition has finished. You can see this at the top right hand corner after you have started the transcription. There, it will say "queue: 1/1" if you are number 1 in line. This means that someone else's speech recognition is running. If it says "queue: 2/2," you are second in line and so on.
To use the link function, do as follows:
You can paste a link to an audio or video file into the URL field. Such links often end with .mp3 or .wmv or similar.
If you want the tool to transcribe directly from an audio or video player in a browser, you can open the browser's developer tools (often by pressing F12). Go to the "Network" or "net" tab. As you load the page, you'll see some links that end with "m3u8." Click on them, copy the link, and paste it into the input field in the speech recognition page. This can work for pages like kvf.fo and logting.fo.
